Language Selection

English French German Italian Portuguese Spanish

Linux: Tuning CFS

Filed under

Nick Piggin used 'git bisect' to track a lmbench regression to the main CFS commit, leading to an interesting discussion between Nick and Ingo Molnar. Ultimately the regression was tracked down to the temporary configurability of the scheduler while it is tuned for optimal performance, "one reason for the extra overhead is the current tunability of CFS, but that is not fundamental, it's caused by the many knobs that CFS has at the moment." The solution, already coded but not yet merged in the mainline kernel "changes those knobs to constants, allowing the compiler to optimize the math better and reduce code size," and as a result result, "CFS can be faster at micro-context-switching than 2.6.22."

Ingo described the lmbench configuration in question as a "micro-benchmark", and noted that with a macro-benchmark better performance was more pronounced, "because with CFS the _quality_ of scheduling decisions has increased.

More Here

More in Tux Machines

Type Title Author Replies Last Postsort icon
Story More of That Cow... Roy Schestowitz 26/10/2016 - 8:25pm
Story Tech Webinar: Knox Tizen Wearable SDK – October 27, 2016 Rianne Schestowitz 26/10/2016 - 6:37pm
Story Open Source Operating Systems for IoT Rianne Schestowitz 26/10/2016 - 6:30pm
Story New Cortex-M chips add ARMv8 and TrustZone Rianne Schestowitz 26/10/2016 - 6:24pm
Story OpenStack in the Headlines Roy Schestowitz 26/10/2016 - 6:12pm
Story Alpine Linux 3.4.5 released Roy Schestowitz 26/10/2016 - 6:00pm
Story Linux Graphics Roy Schestowitz 26/10/2016 - 5:27pm
Story Games for GNU/Linux Roy Schestowitz 26/10/2016 - 5:25pm
Story today's leftovers Roy Schestowitz 26/10/2016 - 11:10am
Story Leftovers: Ubuntu and Debian Roy Schestowitz 26/10/2016 - 11:07am